Today at Computex, we’re unveiling the next phase of our Covert Protocol technical collaboration with NVIDIA, with a hybrid deployment that brings NVIDIA Riva Automatic Speech Recognition (ASR) and NVIDIA Audio2Face on-device for an even faster player experience and economical deployment costs.
During GDC 2024, we worked with NVIDIA to develop Covert Protocol, a demo that unlocks social simulation game mechanics using Inworld’s AI agents. We’re excited about the future of on-device generative AI.
Covert Protocol is just one example of how developers can take advantage of hybrid deployments to integrate advanced AI capabilities into their games. As more powerful multimodal and language models become smaller and more efficient, the future of on-device AI for developers feels not just promising, but inevitable.
Learn more about our Covert Protocol collaboration: https://bit.ly/4bFPrut#NVIDIA#Computex#computex2024#nvidiaace#inworld#ainpcs#aiagents#gaming
Do you know what room Martin is staying in? Ohh yes, because I am definitely Martin's assistant and know all of his whereabouts. No idea. Diego is one of the main characters in Covert Protocol, a demo that brings together the best of NVIDIA and in World to unlock social simulation game mechanics. In this video, we'll be diving into how inworld is powering cognition, perception, and behavior for the AI agents in the game. Diego is an executive at Nexa Life. He has a rich back story powered by his personal knowledge as well as common knowledge that he shares with other characters. When you meet Diego for the first time, he is brusque and dismissive. What's your speech about? Not interested in discussing it with someone who isn't a VIP at the conference. Moving on. The player's response can be captured on device and passed to NVIDIA Riva automatic speech recognition and then passed to inworld for inference. Now let's see what happens after the player has obtained the conference badge and pretends to be an attendee. You look sharp. Where'd you get that suit? Thank you, Sir. I have a personal tailor who always makes sure I'm dressed to impress. Behind the scenes, several in world systems and models are working together to orchestrate Diego's response. The players voice input can be captured on device by NVIDIA Riva automatic speech recognition and the text transcript is then passed to inworld for inference. When the player introduces themselves as Alex Miller, the name on the conference attendee badge, this triggers a mutation of Diego's motivations and demeanor, making him more collegial and friendly with the player. Intent based mutations are also used to trigger changes to the player profile from tourist to attendee. To aid and agents cognition. We've added a customized reasoning step to evaluate the interactions and dynamically evaluate motivations. Customize reasoning allows AI agents to simulate human like decision making processes, engaging in complex interactions, pursue their own goals and create dynamic gameplay scenarios. Strict safety filters are enabled for topics like politics and religion, which are not topics. That the characters should comment on every interaction draws on Diego's basic character sheet, which informs his perception of the game world and lore, as well as his core ego and persona in world's cognition and perception. Systems also output character behavior across multiple data streams to orchestrate the full character performance. NVIDIA Audio to Face and NVIDIA Riva automatic speech recognition can be run on device. To augment our characters facial animation pipeline, we pass emotion parameters so developers can sync. Used to client side animations. As Diego warms up to the player, he will lean forward to indicate interest. I have some insider information about one of your competitors I'd love to meet up later and discuss. Hmm, sounds intriguing. But unfortunately my schedule is quite packed with conference events. Maybe we can discuss over dinner. My treat, of course. Voices are synthesized at runtime in the cloud for expressive vocal performances. Covert Protocol is just a first step and one example of how developers can take advantage of hybrid deployments to integrate advanced AI capabilities into their games. We're continuing to make progress on bringing even more powerful language and multimodal models on device, so let us know in the comments how you would take advantage of this functionality as a game developer.
The integration of NVIDIA Riva ASR and Audio2Face on-device opens up new possibilities for immersive gaming experiences by reducing latency and costs. Imagine AI agents that can understand context, adapt in real-time, and offer personalized interactions seamlessly. This shift is reminiscent of how John Carmack envisioned computing power enabling richer virtual worlds—now we see it manifesting through advanced hybrid deployments in gaming.
🚀 Exciting news! NVIDIA ACE is now generally available, empowering developers to create lifelike digital humans with cutting-edge generative AI. With new microservices for natural language understanding, speech synthesis, and facial animation, industries from gaming to healthcare can revolutionize character interactions. Plus, the new ACE PC NIM microservices bring these capabilities to RTX AI PCs and laptops. Dive into the future of digital human technology with NVIDIA ACE today!
#AI#DigitalHumans#NVIDIA#ACE#TechInnovation
Results oriented, great communicator that uses a deep knowledge of entertainment production systems to provide innovation and success, on time and on budget.
#Picaso is going to be the way to go if you can implement the kinds of controls we can get with stable diffusion. I’m confident in that.
Currently one of NVIDIA performance gains is from generating images where only one pixel in every 8 is from the source. Why not take control of the rest.
When I think of Omniverse as a backed to Picasso as a front end… things get really beautiful. Working in simulation is going to revolutionize the way we dream cinema.
I’m beginning to think of #StoryTwin like a BadRobot mystery box. Imagine l speaking into a Walkie talkie and at the other end of that radio is going to be a generative engine that will work with you, the filmmaker, to bring your idea to life in creative real time .
The future is so rad.
Video, Audio and Conversational AI at NVIDIA
4moI love this video so much, amazing!