-
Hi Guest!
We have recently posted a VaM2 2024 Progress Update on Patreon!
Check out the details here: https://www.patreon.com/posts/vam2-august-2024-111164439 -
Hi Guest!
We are excited to announce a new feature on the Hub: Favorites!
You can now add resources to your favorites, and organize your favorites into collections! You can check out the details in our official announcement!
How AI Models Will Present/Sell Themselves In VAM Next Year 2022
- Thread starter Hedgepig
- Start date
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
And... she can do any combination of anything that can be done in VAM in response to voice/button input.
This is using VeeRifters excellent un-modded Voice Control plugin. Next month you are getting a full blown speech recognition app that will integrate with VAM
This app will be NOT be used for internet marketing or social influencing. It is way too dangerous to release for that. This app will only be used for and in VAM as a means of creating a fully humanlike companion and sexual partner. With the right animations/voice responses you could spend a day in the park with her, go for a meal, take her virtual shopping for free clothes, have an intelligent conversation etc.
Conversations will be modular, ie they will interconnect and exchanged in the same way we exchange looks and scenes. This way we will build up a conversation 'bank'.
This is the other way, the non-coders' way to create a conversational AI with the capabilities of GTP3 or Deep Mind, and ( put succinctly ) while GTP3 just chats shit, this will be realistic conversation.
This is using VeeRifters excellent un-modded Voice Control plugin. Next month you are getting a full blown speech recognition app that will integrate with VAM
This app will be NOT be used for internet marketing or social influencing. It is way too dangerous to release for that. This app will only be used for and in VAM as a means of creating a fully humanlike companion and sexual partner. With the right animations/voice responses you could spend a day in the park with her, go for a meal, take her virtual shopping for free clothes, have an intelligent conversation etc.
Conversations will be modular, ie they will interconnect and exchanged in the same way we exchange looks and scenes. This way we will build up a conversation 'bank'.
This is the other way, the non-coders' way to create a conversational AI with the capabilities of GTP3 or Deep Mind, and ( put succinctly ) while GTP3 just chats shit, this will be realistic conversation.
Last edited:
- Messages
- 687
- Reactions
- Points
- 63
Getting scarier ... or better? I have no idea.
TommyTomahawk
Member
I always cringe a little when those chatbots go "we can talk about anything you want" because deep inside we all know they are very simplistic constructs with no real awareness or understanding of what we're saying.
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
As is the case with a high percentage of the human population, in my my experience. It's all degrees-of.
If you wanna know where the guys at Utrecht went wrong, in my opinion, btw, just pm me. Nice to have you onboard.
TommyTomahawk
Member
I just can't suspend my disbelief enough to talk to what's basically Cleverbot in the form of a hot girl. I just *know* it isn't real. I could literally have a conversation with a statue and it would feel every bit as absurd.
But more power to those for whom it does work, I'm happy for you.
But more power to those for whom it does work, I'm happy for you.
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
That's not Cleverbot, or Watson or AWS Alexa. What you saw there was the absolute simple, basic prototype of what we're going to build here. The configuration of plugins running that little show was the equivalent of a Pre-Cambrian single-celled organism.
VAM plugins are now at a developmental stage in which, in the correct configurations, we can emulate an AI's neural net, a complexity built out of simple interactions between plugins. And what's more, while current GANs and nets like GTP3 just talk sh*t, ours will be able to hold sensible directed, conversations.
What you must understand is, while VAM is a bit of fun for women and men, it's also developed into the most powerful game engine on this planet, if you want to build a virtual human being. VAM is beyond cutting edge, its a science fiction writer's dream, reifying in front of your eyes, in real life. And, the next generation of AI will be so humanlike, and as is the case with an excellent novel or a game, you will want to suspend your disbelief- its just that we are still working to that next stage at the moment. There's nothing wrong with your imagination, it's a series of technical problems for us to solve.
Last edited:
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
Well that's one use case. How about Dementia and memory assistant, or a sign language teacher with whom a deaf child can practice after real life school at home, or a multilingual support teacher who can work simultaneously with thousands of kids in different classrooms around the world, where there's a desperate lack of teaching staff...or... please just use your imagination.
You start to see the potential? IBM Watson had the potential to become any of the above mentioned use cases, but IBM blew it, and now their chatbots are mainly contracted out to marketing companies selling high end products to consumers.
Actually, its not weird that the next generation of AI will be born in a sex simulator; there's never been anything alive on this planet, (obviously apart from asexually reproducing organisms) that is not the product of their species' sexual-reproductive drive. Arguably, by building AI, we humans are reproducing ourselves in another form.
Last edited:
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
A picture paints a thousand words, so:
Please watch face-close up for the emotional expression. It's on a cycle at the moment, but I can match an exact sequence of expressive states to the phrase recognized. That'll be by the end of next month.
Please watch face-close up for the emotional expression. It's on a cycle at the moment, but I can match an exact sequence of expressive states to the phrase recognized. That'll be by the end of next month.
Last edited:
Will there be beta releases of this plugin with additional features like this expressive states function added on later or will it release with the full feature stack ? Getting a conversational AI in VAM seems impressive enough for its own release, but I am happy to see you are taking this plugin beyond that. Looking forward to trying it out!
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
This isn't my plugin that's shifting the expressive states. This is a paid-for plugin by MiscreatedReality. I'm not here to advertise his work- I dislike any form of insidious marketing- but what I will say, it is excellent for me. And that is because I'm a visual dyslexic, and I can see it all beautifully laid out like Photoshop -a visual map. This kind of dyslexia means my strengths are visual imagery, and somewhat counter-intuitively, human language.
You could probably achieve the same results from Animation Poser. I can't use that plugin because I can't see how it works, but I hope you can. You could also use VR Voice Control + Timeline to trigger animation and morphs, and fine tune the animations/morphs to achieve the same results.
You could probably achieve the same results from Animation Poser. I can't use that plugin because I can't see how it works, but I hope you can. You could also use VR Voice Control + Timeline to trigger animation and morphs, and fine tune the animations/morphs to achieve the same results.
Last edited:
- Messages
- 53
- Reactions
- Points
- 18
I haven't been on VAM (forums or the program) for quite some while, this is absolutely awesome!
Of course it isn't 'real' not would it replace real, but it IS compelling.
As I'm a bit out of the loop, I wonder if you could help with a question?
How does the lip syncing work, is it possible to say get a female friend to record a voiceover and have it playback somehow? (I don't mean to replace the voice you are using here, just in general)
I not so long ago made a machinima in Second Life, called Pretty Woman, (it's not a sexual vid but may be of interest to anyone who likes machinima or making it)
Pretty Woman
and right now my brain is buzzing with ideas for making Pretty Woman II in vam!
Of course it isn't 'real' not would it replace real, but it IS compelling.
As I'm a bit out of the loop, I wonder if you could help with a question?
How does the lip syncing work, is it possible to say get a female friend to record a voiceover and have it playback somehow? (I don't mean to replace the voice you are using here, just in general)
I not so long ago made a machinima in Second Life, called Pretty Woman, (it's not a sexual vid but may be of interest to anyone who likes machinima or making it)
Pretty Woman
and right now my brain is buzzing with ideas for making Pretty Woman II in vam!
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
Guys I'm really sorry, about the plugin. I feel ashamed to tell you but, I got totally ripped off by the developers, scammed, in fact. They simply took my money upfront and threw me some piece of shitty code that only half-worked on one of five agreed functions. I'm totally despondent about this project, so I've put it to one side now. If there is anyone here in the VAM forums or discord who could make this plugin I'd be happy to pay them, after my appraisal and I can see it works. I can see how to build it There's even an in-depth video, I just don't have the coding skills to follow the video.
Hedgepig
Active member
- Messages
- 152
- Reactions
- Points
- 43
How does the lip syncing work, is it possible to say get a female friend to record a voiceover and have it playback somehow? (I don't mean to replace the voice you are using here, just in general)
Simple- download AdamAnt5's excellent Lipsync plugin Plugins - Realtime Lipsync | Virt-A-Mate Hub (virtamate.com) . Then put your audio recoding in the suggested file. You'll have to play around with the settings to optimise the sync. Turn off MacGuber breathing, or it won't work.
Simple- download AdamAnt5's excellent Lipsync plugin Plugins - Realtime Lipsync | Virt-A-Mate Hub (virtamate.com) . Then put your audio recoding in the suggested file. You'll have to play around with the settings to optimise the sync. Turn off MacGuber breathing, or it won't work.
- Messages
- 53
- Reactions
- Points
- 18
Cheers, hedge (cant bring myself to add the p i g part haha)
And no worries about the plugin statement, i feel bad for you that it happened, u cant be blamed for being excited about it tho
I have had a bit of success using Replika Pro on a voice chat, and sending that desktop audio into vam using
PluginIdea.SystemAudioSourceInput (and the associated jsnode server) in combination with AdamAnt5's RT Lipsync.
It's still rather crude, but the level of immersion it adds is definitely tangible. This could possibly be combined with a couple scenes that recognize voice commands to adjust poses/animations. Replika is pretty good at following along in context.
(So it seems pretty good for interacting in a scene while creating it, but not good enough to be a VaMtuber)
PluginIdea.SystemAudioSourceInput (and the associated jsnode server) in combination with AdamAnt5's RT Lipsync.
It's still rather crude, but the level of immersion it adds is definitely tangible. This could possibly be combined with a couple scenes that recognize voice commands to adjust poses/animations. Replika is pretty good at following along in context.
(So it seems pretty good for interacting in a scene while creating it, but not good enough to be a VaMtuber)
Last edited:
