Featured

Introducing: The world's fastest Conversational Video Interface for developers

Julia Szatar
August 15, 2024
min read
Contributors
Build AI video with Tavus APIs
Get Started Free
Share

At Tavus, our mission is to make digital experiences as immersive as human face-to-face interactions by empowering people to leverage their likeness at scale online. 

Back in March, we launched our breakthrough Digital Replica model, Phoenix, and Video Generation on our developer platform. 

Today, we’re thrilled to announce: the Conversational Video Interface. Developers can now build rich, realistic, real-time conversational experiences with digital twins on the Tavus platform. 

Try talking to Carter in our live demo on our homepage.

Try a live demo on www.tavus.io

A new human-computer interface

The Conversational Video Interface (CVI) is the only solution on the market that gives developers a complete set of building blocks to create interactive experiences with digital twins that speak, see, and hear. 

We’ve delivered a conversational suite that stands apart from the rest. 

  • The world’s fastest: with less than one second of latency between utterances
  • The only end-to-end solution: deploys easily without any deep eng work
  • The most realistic: with a natural conversational cadence and our replica model Phoenix-2

Developers in industries like the creator economy, education, eCommerce, and sales are already building with the Tavus CVI to scale human abilities and reinvent how we interact in the digital realm.

Users can talk to digital twins that speak, see, and hear.

Why AI powered conversational video?

Historically, technology allowed us to scale communication across geography, time, and people. We started with letters and carrier pigeons, then we got the telephone, and later television. Then came the internet and eventually video conferencing. 

Throughout this evolution we’ve had to adapt to technological limitations which often forced us to lose a touch of our humanity. And, if we focused on a personalized touch, we had to trade off on scale.

The beauty of AI video is that now technology can meet us where we naturally communicate, while maintaining unprecedented scalability.

One-on-one mentorship is revolutionized with digital cloning

Last week, our customer Delphi, the personalized mentorship and education platform, announced its groundbreaking Video Clone feature. Enabled by Tavus’ technology, this feature allows real-time video interactions with digital clones of creators, experts, coaches, and executives, providing a personal mentor on demand.

“There are a lot of components within a conversation. It’s incredibly complicated for an AI system to power a Digital Clone that can carry on a natural, live conversation over video,” said Dara Ladjevardian, Co-Founder and CEO of Delphi. 

“Tavus tackles this challenge beautifully. We chose to partner with them because they have developed the world’s first conversational solution with under a second of latency. Their research and technology delivers an incredibly realistic interactive experience. This is critical to our ability to deliver authentic and credible personalized mentorship experiences with expert clones on our platform.”

Features and functionality highlights

Here’s why you should build AI agents with the Conversational Video Interface

End-to-end: Get started immediately with pre-built end-to-end components.

  • Build safe digital twins and stock AI agents with the replica API
  • Customize the LLM, persona, memories, context, and scenario for conversations
  • Launch and stream human-to-AI conversations in an embeddable meeting rooms powered by Daily
  • Record, transcribe, and share the conversation
  • Handle high traffic with ease with production-grade scalability

Realistic: Our CVI delivers the most realistic white-labeled video interactions on the market. 

  • Lowest latency between utterances on the market at one second
  • Hyper-real digital twins with state-of-the-art cloning 
  • Near-instant boot time
  • Rolling vision, interruptibility, and end-of-turn detection
  • A purpose-built conversational pipeline and fine tuned LLM

Modular: We built our solution with developers in mind using customizable components.

  • Choose digital twins or stock replicas
  • Easily connect your own LLM, or models like GPT-4o and Claude
  • Swap our TTS for your preferred solution 
  • Use our real time replica, and bring your own streamed in audio or text, if preferred

See developer docs.

Will you build digital twins or AI agents?

For the longest time, technology has pushed human interaction towards the transactional. Now, AI video can apply a human touch at scale in any industry.

We see two distinct directions for using CVI to build real-time AI-powered interactions:

Digital Twins: Extend the presence of high-impact individuals with specialist knowledge, such as executives, experts, coaches, professors, healthcare professionals, and celebrities, to overcome limitations of time, scale, and knowledge.

AI Agents: Place intelligent AI agents with a face, a voice, warmth, and humanity, where leveraging humans is not feasible today. Examples include customer support agents, digital sales assistants, personal assistants, and technical co-pilots across industries like eCommerce, government services, education, software, and entertainment.

Sign up for free

We aim to revolutionize the way people interact and work in the digital age, ushering in a future where the boundaries between human and machine capabilities are seamlessly and safely integrated.

We’re so excited to see how developers leverage CVI to build AI-powered conversations that expand human abilities across use cases and industries.

If you have an idea in mind, sign up for free to test our APIs and suite.

Research initiatives

The team is at the forefront of AI video research and pushes model updates every two weeks based on the latest research and customer needs.

Industry
min read
This is some text inside of a div block.
min read

Synthesia API Review & Alternatives for AI Video Generation [2024]

Explore Synthesia API and its 2024 alternatives. Learn about each tool's features, weigh their pros and cons, and find the right API video solution for you.
Industry
min read
This is some text inside of a div block.
min read

15 Best Voice Cloning APIs | 2024

Increasingly realistic voice cloning APIs can now help businesses create content at scale. We’ll explore the capabilities of voice cloning and the top APIs in 2024.
Industry
min read
This is some text inside of a div block.
min read

What is a Stock Avatar? | 2024

It can be confusing to know the differences between stock avatars and other types of virtual humans. Learn what a stock avatar is and does, and its benefits.
Product
5
min read
This is some text inside of a div block.
min read

How to do Text to Video for AI Replicas

An introduction to how to make an AI video with Tavus' video API.
Product
min read
This is some text inside of a div block.
min read

Build Your First Real Time Conversational Digital Twin in Five Minutes

A tutorial on how you can build a conversational AI in just a few minutes with details of what each of the parameters do.
Product
5
min read
This is some text inside of a div block.
min read

Build a Custom Personality for Real Time Video AI

Customize your conversational AI's personality using a custom persona with system prompts, context, and your own LLM.

AI video APIs for digital twins

Build immersive AI-generated video experiences in your application