L2TOR · Project

Social Robots That Teach Young Children a Second Language Through Interactive Play

digitalPilotedTRL 6

Imagine a friendly robot that sits with a 4-year-old and teaches them English the way a patient tutor would — using gestures, pictures, and conversation adapted to the child's level. That's what this project built. The robot picks up on how the child is doing and adjusts its teaching style in real time, something that screen-based apps can't do nearly as well. They tested it with children in the Netherlands, Germany, and Turkey learning English, and also helped immigrant children pick up the local language.

By the numbers

3,042,562

EUR in EU funding for development

consortium partners across 6 countries

total project deliverables completed

year-old target age for language learners

industry partners in the consortium

The business problem

What needed solving

Most children in Europe need to learn a second language, but there aren't enough qualified native-speaking tutors to provide the one-on-one attention that young children need. Screen-based apps fall short because 4-year-olds learn best through physical, social interaction — gestures, eye contact, and adaptive conversation. Schools and EdTech companies need a scalable way to deliver personalized language tutoring to very young children.

The solution

What was built

The project built a social robot tutoring system with multimodal interaction (speech, gestures, social signal processing) designed specifically for 4-year-old children learning a second language. A demonstrator system was deployed in associated schools across multiple countries, with 28 deliverables covering the technical platform, pedagogical methods, and child interaction studies.

Audience

Who needs this

EdTech companies building language learning products for children under 6Social robotics manufacturers looking for proven educational applicationsInternational preschool and kindergarten chains serving multilingual populationsGovernment education departments seeking scalable language integration toolsPublishers of early childhood educational content exploring interactive formats

Business applications

Who can put this to work

EdTech and Language Learning

SME

Target: Companies developing language learning products for children (apps, platforms, content)

If you are an EdTech company building language learning tools for young children — this project developed a robot-based tutoring system that outperforms screen-based alternatives in motivation and learning outcomes. The system was tested with 4-year-olds learning English across 3 countries. You could license the interaction design and pedagogical methods to add a physical robot companion to your product line.

Social Robotics and Consumer Electronics

mid-size

Target: Robotics manufacturers producing educational or companion robots

If you are a robotics company looking to expand into the education market — this project built multimodal interaction technology including social signal processing and human-robot interaction management specifically for child users. The system was deployed in associated schools as a demonstrator. You could integrate these interaction capabilities into your existing robot platforms to enter the language tutoring market.

Early Childhood Education

any

Target: Childcare chains, international preschool networks, and school systems

If you run preschools or early education centers with multilingual student populations — this project created a robot tutor deployed in schools that helps 4-year-old children learn a second language through personalized, interactive sessions. With 7 partners across 6 countries validating the approach, the system addresses the shortage of native-speaking tutors in early childhood education.

Frequently asked

Quick answers

What would it cost to implement robot tutors in our schools or products?

The project received EUR 3,042,562 in EU funding to develop and test the system across 7 partner institutions. Commercial pricing would depend on the robotics hardware platform and licensing terms. Based on available project data, no per-unit cost figures are provided.

Can this scale beyond a classroom demo to hundreds of schools?

The project produced a demonstrator system deployed in associated schools, proving the concept works in real educational settings. Scaling would require productizing the software and pairing it with commercially available social robot hardware. The multimodal interaction system was designed to work with children speaking Dutch, German, Turkish, and English.

Who owns the intellectual property and can we license it?

The consortium of 7 partners across 6 countries jointly developed the technology, with the University of Plymouth as coordinator. IP rights are governed by the EU grant agreement. Licensing discussions would need to go through the consortium, likely starting with the coordinator.

Has this actually been tested with real children in real classrooms?

Yes. The project includes a demonstrator deliverable specifically described as deployment in associated schools. The system was tested with 4-year-old children learning English as a second language in the Netherlands, Germany, and Turkey.

How does this compare to tablet-based language apps?

The project objective states that social robots have marked benefits over screen-based tutoring technologies, with demonstrable positive impacts on motivation and learning outcomes. The robot uses multimodal interaction — gestures, speech, and social signals — which tablets cannot replicate.

What languages does it support?

The system was built for Dutch, German, and Turkish children learning English. Additionally, it supports Turkish immigrant children in the Netherlands and Germany learning Dutch and German. Expanding to other language pairs would require additional development.

Is this ready to buy off the shelf?

Not yet. This was a research and innovation project (2016-2018) that produced a working demonstrator deployed in schools. Moving to a commercial product would require engineering for reliability, content expansion, and hardware partnerships. The core technology and pedagogical methods are proven.

Consortium

Who built it

The L2TOR consortium brings together 7 partners from 6 countries (Belgium, Germany, France, Netherlands, Turkey, UK), led by the University of Plymouth. The mix is research-heavy with 5 universities and 2 industry partners (29% industry ratio), which is typical for a technology at this stage — the science needed to be proven before commercial partners would invest heavily. The multinational spread is a strength for language learning: they had native environments for Dutch, German, Turkish, and English testing. For a business looking to commercialize, the 2 industry partners are the most likely entry points for licensing or partnership discussions.

UNIVERSITY OF PLYMOUTHCoordinator · UK
KOC UNIVERSITYparticipant · TR
TILBURG UNIVERSITY- UNIVERSITEIT VAN TILBURGparticipant · NL
UNIVERSITEIT UTRECHTparticipant · NL
UNIVERSITAET BIELEFELDparticipant · DE
ALDEBARANparticipant · FR

How to reach the team

University of Plymouth, UK — reach out to the project coordinator through the university's research office or the project website

Order a report on this consortium See UNIVERSITY OF PLYMOUTH profile →

Next steps

Talk to the team behind this work.

Want to explore licensing this robot tutoring technology for your EdTech product or school network? SciTransfer can connect you with the right people in the consortium.

Order a report on this topic More in Digital & ICT