Estonian scientists working on first sign language translation bot

A unique project has been launched in Estonia to create the first Estonian sign language translation robot.
The solution under development takes the form of a video avatar that imitates signing. It is designed to assist the sign language community, particularly in everyday situations where interpreting services are currently unavailable or difficult to access. At first, the team will focus on translating transportation announcements, news and museum exhibit descriptions as part of building the prototype.
For many deaf people, Estonian Sign Language is their native language, while written Estonian functions as a foreign language. Therefore, simply relaying spoken information is not enough — translation from text into sign language is also needed.
Sirle Papp, chair of the Estonian Association of the Deaf, called the initiative highly important. "The idea of developing a signing robot is both commendable and necessary because it helps bring everyday information closer to people — whether in public institutions, shopping and activity centers or transport hubs," Papp said. She added that such a solution supports the creation of a more accessible society and increases opportunities for deaf individuals to obtain information independently.
According to Mark Fišel, professor of language technology at the University of Tartu's Institute of Computer Science, the project also poses a significant scientific challenge. He believes that the AI-based translation robot will lay the foundation for an entirely new field of research in Estonia.
For the Institute of the Estonian Language, the project is part of a broader mission. Kadri Vare, head of the institute's language technology department, said their goal is to ensure that Estonian and Estonian Sign Language evolve hand in hand with the digital age. "We are not just talking about technology, but about access and equal opportunities," Vare noted.
The results will be tested in close cooperation with the deaf community to ensure the solution's practical value and user-friendliness. The team hopes the initiative will grow into a new scientific discipline that bridges language technology and sign language research in Estonia.
The project is led by the Institute of the Estonian Language, with partners including the University of Tartu's Institute of Computer Science and the technology companies STACC and Texta.
--
Editor: Marcus Turovski










