To interact with us robots need to see, hear, speak, and express themselves as naturally as we can. Furhat is specifically designed for human social interactions using a multimodal system with discrete modular subsystems that handle functions such as facial animations, neck motion, visual perception, audio processing, cloud service integrations and other operations that allow it to interact with humans just as we interact with each other.
The powerful combination of back-projected facial animations, swapable masks and face editing tools make it very easy to create expressive robot characters with any ethnicity, age and gender. Furhat also comes with built-in gestures such as blinking, smiling, nodding, frowning, and other gestures that can be controlled individually or fully automated with Large Language Models.
Furhat can see, hear and track multiple users simultaneously in real-time, do facial expression analysis, estimate head pose and user distance, giving it the ability to have unique interactions with each individual.
Furhat is designed for natural human conversation with rapid turn-taking in multiple languages. You have detailed controls for managing initiative, turn-taking, interruptions, error handling and priming the speech recogniser for expected utterances. Furhat is also integrated with leading speech recognition and voice synthesis providers including Microsoft Azure, Amazon Polly and Elevenlabs with over 200 voices available in 120 languages and dialects.
Want to get a detailed overview of Furhat's modular system? View or download the technical specifications below.
Build interactions in Kotlin or use other programming languages with the Remote API. Furhat comes with a powerful set of programming tools for researchers, educators, developers, and students.
Explore SDKDon't have any prior experience with coding? You can also use our LLM-driven conversation designer to rapidly ideate, create, and test interactions through prompting.
Explore Creator