Get updates delivered to you daily. Free and customizable.
Interesting Engineering
US: World’s fastest digital twin humanizes AI for real-time video chats
By Abhishek Bhardwaj,
1 day ago
A generative artificial intelligence (AI) video research company has announced the release of the world’s fastest digital twin solution for real-time video conversations.
Tavus, the San Francisco based firm, has named the new product, Conversational Video Interface (CVI).
According to the firm, the new product gives developers a complete set of building blocks which they can use to create interactive experiences with digital twins that can see, hear, and respond with a minimal lag.
Tavus claims that the digital twins made through the platform can act almost at par with the humans they represent, which can possibly make the lives of content creators, help desk personnel, and others a lot easier.
“Conversational replicas represent a paradigm shift in the way we interact with machines,” said Hassaan Raza, Co-founder and CEO of Tavus. “Rather than adapting to the limitations of traditional interfaces, we are creating solutions that leverage the natural ways humans communicate – through visual and verbal cues, emotional understanding, and contextual awareness.
Tavus’ new Conversational Video Interface
Tavus claims that its CVI creates a lifelike digital twin that goes beyond simple chatbots or pre-recorded videos.
Leveraging AI, the product creates a dynamic replica that embodies the person’s likeness, knowledge, context, and background, as per the company. It can see and process visual and voice information, answer questions, provide insights, and engage in conversations, just as the real individual would.
“The replica draws upon a person or company’s expertise, experiences, and unique perspective to deliver authentic, personalized responses, essentially acting as a proxy,” the company says.
According to a press release , its new CVI can communicate with less than one second of latency during conversations and when asked to carry out any specific task.
Moreover, the company also says that it can be deployed without much critical work on the user’s part as it is an end-to-end developer solution. Other features which come with Tavus’ new offering are rolling vision and a conversational pipeline which is built to suit different purposes.
Other features and security measures
It also offers the room to experiment as it can also be paired with models like ChatGPT-4o or Claude, or any other large language model (LLM).
The platform also offers production scalability and parallel stream management options to handle high traffic volumes.
Coming to the safety side of things, the company says it has installed safeguards to ensure that an individual can only make a digital twin of themselves. It also boasts of SOC 2 compliance, automated content moderation and anti-hallucination quality checks.
“This new video interface will enable more intuitive, efficient, and impactful interactions. In short, CVI allows users to leverage and scale their digital likeness, creating a true extension of themselves that overcomes the limitations of space, time, and memory. It also opens new doors for developers to create unparalleled customer experiences,” Raza said.
However, while generative AI is progressing by leaps and bounds there are several apprehensions that society has regarding safety and its possible misuse.
The world of AI is exploding fast and more such products are going to appear in the days ahead which will require due diligence and more vigilance to minimize chances of online scams, unauthorized impersonation, and more.
Get updates delivered to you daily. Free and customizable.
It’s essential to note our commitment to transparency:
Our Terms of Use acknowledge that our services may not always be error-free, and our Community Standards emphasize our discretion in enforcing policies. As a platform hosting over 100,000 pieces of content published daily, we cannot pre-vet content, but we strive to foster a dynamic environment for free expression and robust discourse through safety guardrails of human and AI moderation.
Comments / 0