Our platform is about people.
And it turns out, fostering natural interactions with avatars isn’t easy. It takes a lot of science and technology to make a voice match its digital face.
How Rapport was built
Under the hood: actual science.
Our core technology is based on 10+ years of scientific research in linguistics, biomechanics, psychology, machine learning, and computer graphics.
Real-time, right when you need it.
We designed our cloud-based platform with your real-world applications in mind. Speech parsing, analysis, and synthesis happen in real-time. No more lagging responses and out-of-sync animation.
Easy to start, use, and integrate.
Easily integrate our API into your own tech with a few lines of code, or with other APIs like speech-to-text, text-to-speech, and conversational AIs.Try it out
How it all works
Rapport’s Speech Graphics engine takes a voice stream and analyzes the spoken and unspoken sounds in the audio to capture the emotions of the speaker and generate the metrics to animate a digital avatar. The engine can coordinate data streams with external speech-to-text, text-to-speech, and conversational AIs to create seamless, low-latency conversations in real-time.
Where’s the science?
There are hundreds of muscles in the face, tongue, and neck responsible for the thousands of small movements making up human speech and nonverbal expression. Understanding and controlling how these muscles are used during communication is critical to making our animation realistic, and bringing avatars to life.
Our team of linguistic scientists help us understand the relationship between the sounds we make and the facial movements required to make them—in any language.
In addition to our cutting edge research, we also construct and train our machine learning models to help analyze sound and turn it into animation in real-time across a global network.
Where’s the science?
Join the waitlist and get early access to our platform.
We’re giving early access to the kind of people who like pushing the envelope. People like us, basically. Sign up to be one of the first to build the future with our technology.Get early access