When the founder of Soul, Zhang Lu introduced the app to the world in 2016, it was obvious that this wasn’t a run-of-the-mill social networking platform. The app was designed for people who sought connections exclusively based on interests, passions, and inclinations.
As such, Soul Zhang Lu pioneered the idea of using not photographs but avatars of the user in digital interactions. Unlike on other platforms, avatar not just meant to be a tiny, inconsequential aspect of user’s digital personality. When using Soul, avatar the digital personality.
So, the Soul team offer a variety of ways in which these avatars could be designed to represent various aspects of the user’s personality. Introduced in 2022, NAWA allows Soul users the ability to create their own avatars.
And, Soul Zhang Lu has encouraged the platform’s engineers to continue working on abilities. The progress made in this direction was presented by the company at the recently concluded GITEX GLOBAL held in Dubai.
The event which provides a platform for the latest technological innovations that drive change across industries saw a massive turnout of over 6,700 companies as well as tech stalwarts and enthusiasts. So, the engineers at Soul had the perfect audience for the debut of their latest advancement which marks a significant milestone in the intersection of AI, social networking, and digital companionship.
Hitherto, avatars have always played a static role in digital interactions. Even 3D avatars would simply depict one aspect of the user’s personality or mood. There was no way for these digital displays to mimic the user behind the screen. But, Soul Zhang Lu’s team changed that, and in a big way too.
For starters, monocular cameras, the kind found in most smartphones, were used for motion capture. The data gathered was then seamlessly provided to the engine in real-time for conversion to various aspects of a person’s avatar. GITEX GLOBAL attendees were left stunned as their 3D avatars moved about and expressed themselves.
In terms of application, this extraordinary ability of avatars to pick up and convey human sentiments in real time offers a more fulfilling and emotionally resonant interaction experience. As a matter of fact, it could be said that the experience is so immersive that it feels astonishingly close to real life; almost like having a face-to-face conversation with another person.
So, it would be safe to say that Soul App’s multimodal AI model is capable of redefining social connection through a new level of interactivity. In general, it reflects a broader trend in the tech industry toward creating immersive digital experiences that are lifelike.
The team of Soul Zhang Lu calls these 3D avatars, “Digital Twins” and rightly so. By combining natural language processing, real-time motion capture, and voice recognition, Soul’s model gives these virtual avatars a remarkable ability to accurately understand and replicate human communication nuances. The level of personalization that these “Digital Twins” bring to an interaction is simply unparalleled.
In terms of just how much can be achieved with this new multimodal AI model, at this time, theengine is capable of recognizing over 90 different facial expressions and gestures and using them to adapt the responses of the avatars to create lifelike interactions.
A vital aspect of the behavior of Soul Zhang Lu’s Digital Twins is that they are not just able to replicate human motions but also personality traits. This brings a whole new level of realism to the interactions.
Also, the avatars can convey emotions and foster interactions that feel genuine and human. This approach to social networking resonates particularly well with younger audiences, who are increasingly seeking digital platforms where they can connect authentically. Since Soul Zhang Lu’s platform enjoys exceptional popularity among China’s Gen Z, this upgrade is bound to get the platform’s users hopping excited.
In terms of avatar creation, Soul Zhang Lu’s team has made the entire process pretty simple. The powerful engine takes but a few seconds to create realistic avatars to suit the user’s preferences and instructions. Users can rely on text instructions, chat history, and even photographs to generate the avatar. From thereon, users can personalize the avatars by changing their body language, voice tone, and personas.
Once created, the avatars can engage in seamless multimodal interactions that combine text, voice, and movement. The level of interaction these avatars offer breaks the “dimensional barrier,” allowing users to socialize in a way that is immersive and lifelike, even within a digital environment. In other words, Soul has managed to blend the real with the virtual such that the former powers the latter.
As far as advancements on this front are concerned, future updates will focus on enhancing the avatars’ ability to recognize and interpret emotions, providing a more personalized experience that mirrors the complexities of human relationships.
