Soul App has recently made a major breakthrough in the field of "AI + social"! The editor of Downcodes learned that Soul officially announced that its self-developed end-to-end full-duplex voice call model has been upgraded again, enabling natural and smooth voice conversations with virtual people as natural and smooth as real people. This move marks an important step for Soul in the application of AI technology, bringing users a more immersive and interactive social experience. This article will delve into the unique features of this large model and Soul’s exploration in the AI social field.
On the domestic "AI + social" track, Soul App is about to use AI to inject new vitality!
Recently, Soul officially announced that its voice model has been upgraded again, and a self-developed end-to-end full-duplex voice call model has been launched.
The most amazing effect of this upgrade is that the voice call between the user and the virtual person can be as natural and smooth as chatting with a real person!
How realistic the effect is? You can first watch the video below to get a feel for it:
An official example of "Experience real-time calls with AI"
So what’s so special about Soul’s self-developed end-to-end voice call model? According to the official description, its biggest highlights include:
With ultra-low interaction latency
Quick automatic interruption
Super realistic voice expression
Emotional perception and understanding ability, etc.
The ultra-low interaction delay capability means that the moment you speak, the AI can respond immediately without any delay, and the distance between you and the AI can be shortened in an instant. If you want to have real communication with it, you don’t need to wait at all, it’s just like talking to a real person.
Soul's large voice model supports fast automatic interruption . In other words, when you are communicating with AI, if you want to interrupt, it can completely understand what you mean and interrupt the other party easily. This kind of interaction is really interesting!
Finally, coupled with ultra-realistic voice expression and emotional perception and understanding capabilities , AI can not only understand your words, but also sense your emotions and give appropriate responses based on your emotions.
Based on the official video example, if this feature is fully launched in the future, it is estimated that a large number of users may not be able to distinguish between real people and AI virtual people when they experience it on Soul.
Soul said that its end-to-end voice call large model has been applied to the "Echo of Another World" real-time call scenario (under internal testing), and will be expanded to multiple AI companionship and AI interaction scenarios such as AI Gou Dan in the future.
It is understood that as early as 2020, Soul has launched AIGC technology research and development, focusing on the research and development of key technologies such as intelligent dialogue, voice technology, and virtual humans, and deeply integrating these AI capabilities into social scenarios.
In the process of using AI to upgrade social interaction, Soul pays special attention to achieving an anthropomorphic and natural emotional companionship experience.
In order to bring better emotional feedback and companionship to users, the Soul technical team has been paying attention to emotional understanding and delay issues. They have launched self-developed speech generation models, speech recognition models, voice dialogue models, music generation models, etc., which support real tone generation, voice DIY, multi-language switching, multi-emotional immersive real-time dialogue and other functions. These have already It has been used in multiple scenarios of Soul, such as "AI Goudan", "Werewolf Phantom" AI voice real-time interaction, "Echo from Another World", etc.
Soul's self-developed end-to-end voice call model is now online, which means users can enjoy a more natural human-computer interaction experience. In the future, Soul also plans to further promote the construction of multi-modal end-to-end large model capabilities to make the interaction between people and AI more interesting and immersive.
Soul’s AI technology upgrade this time not only improves the user experience, but also provides new ideas for the future development direction of “AI + social”. I believe that in the near future, we will see the emergence of more innovative social applications based on AI technology, bringing more fun and convenience to people’s social lives.