e-LADDA · Project

Speech Recognition and Apps That Help Young Children Learn Language in a Digital World

digitalPrototypeTRL 4Thin data (2/5)

Kids today grow up swiping tablets before they can tie their shoes, but nobody really knows if all that screen time helps or hurts their ability to talk and learn language. A team of researchers across 8 countries studied exactly that — what makes a digital tool good or bad for a toddler's language growth. They built apps, digital content, and even a prototype that can recognize how young children speak (which is very different from adult speech). Think of it as finally getting a scientific answer to every parent's question: "Is this app actually teaching my kid anything?"

By the numbers

research partners in consortium

countries represented

industry partners involved

total project deliverables

25%

industry ratio in consortium

The business problem

What needed solving

Most speech recognition technology is built for adults and fails badly when young children speak — their pronunciation, vocabulary, and sentence structure are fundamentally different. At the same time, the children's EdTech market is flooded with apps that claim to be educational but have no scientific evidence backing those claims. Companies in both spaces need research-validated tools and guidelines to build products that actually work for young learners.

The solution

What was built

The project delivered technological solutions including educational apps, digital content designed for young children, and a child speech recognition prototype. In total, 9 deliverables were produced across the consortium of 12 partners.

Audience

Who needs this

EdTech companies building learning apps for children aged 0-6Speech recognition companies wanting to improve accuracy for children's voicesChildren's media publishers creating digital contentToy manufacturers integrating voice interaction into productsSchool districts and early education providers evaluating digital tools

Business applications

Who can put this to work

EdTech / Children's Learning Apps

SME

Target: Companies developing educational apps or digital content for children aged 0-6

If you are an EdTech company building learning apps for young children and struggling with engagement or proven educational outcomes — this project developed research-backed guidelines and digital content prototypes showing exactly which features in apps advance language learning and which impede it. With 12 research partners across 8 countries validating results, you get cross-cultural evidence to back your product claims.

Speech Technology / Voice AI

any

Target: Companies developing speech recognition engines or voice assistants

If you are a speech technology company whose recognition engine fails on children's voices — this project built a child speech recognition prototype specifically designed for young learners. Standard speech engines are trained on adult voices and perform poorly with children. The e-LADDA prototype addresses this gap with research from 9 universities specializing in language development.

Children's Media & Publishing

mid-size

Target: Publishers and media companies producing digital content for preschoolers

If you are a children's media company trying to differentiate your digital products with evidence-based claims — this project produced research across 8 countries on what actually works in digital environments for language learning. Their findings on communication channels and technology factors that advance or impede learning can directly inform your content design and give you a defensible marketing edge.

Frequently asked

Quick answers

What would it cost to license or use these technologies?

The project was an MSCA training network coordinated by NTNU in Norway with 3 industry partners already involved. Licensing terms for the child speech recognition prototype and apps would need to be negotiated directly with the consortium. Based on available project data, no commercial pricing has been published.

Can the child speech recognition prototype work at industrial scale?

The deliverable is described as a prototype, not a production-ready system. Scaling it would require additional engineering, training data, and infrastructure investment. The 3 industry partners in the consortium may have insights on commercialization pathways.

Who owns the intellectual property?

IP from MSCA-ITN projects typically stays with the institutions that generated it, following EU grant agreement rules. With 12 partners across 8 countries, IP rights may be distributed. Contact NTNU (coordinator) for specific licensing arrangements.

Does this work across different languages?

The consortium spans 8 countries (Germany, Spain, Latvia, Netherlands, Norway, Portugal, Sweden, UK), strongly suggesting multilingual research coverage. This cross-linguistic scope is a significant advantage for companies operating in multiple European markets.

What concrete outputs came from this project?

The project produced 9 deliverables total, including 1 demonstrated set of technological solutions: apps, digital content, and a child speech recognition prototype. These represent tangible tools that could be adapted for commercial use.

Is there regulatory relevance?

With increasing EU attention on children's digital safety and age-appropriate design (e.g., Digital Services Act), evidence-based guidelines from this project on how digital tools affect young children's development could help companies demonstrate compliance and responsible design.

Consortium

Who built it

The e-LADDA consortium brings together 12 partners from 8 European countries, with a 25% industry ratio (3 out of 12 partners). The coordinator is NTNU, one of Norway's leading technical universities. The consortium is university-heavy (9 out of 12 partners), which is expected for a Marie Curie training network where researcher education is the primary goal. The 3 industry partners — all classified as SMEs — indicate that smaller, agile companies in EdTech or speech technology were involved from the start, potentially making them natural channels for commercializing the research outputs. The geographic spread across Germany, Spain, Latvia, Netherlands, Norway, Portugal, Sweden, and the UK ensures the research has cross-cultural validity, which is essential for any company targeting the pan-European children's digital learning market.

NORGES TEKNISK-NATURVITENSKAPELIGE UNIVERSITET NTNUCoordinator · NO
UNIVERSITY OF EAST ANGLIAparticipant · UK
FURHAT ROBOTICS ABparticipant · SE
KUNGLIGA TEKNISKA HOEGSKOLANparticipant · SE
THE UNIVERSITY OF MANCHESTERparticipant · UK
GEORG-AUGUST-UNIVERSITAT GOTTINGEN STIFTUNG OFFENTLICHEN RECHTSparticipant · DE
UNIVERSITY OF LANCASTERparticipant · UK
UNIVERSIDADE DO PORTOparticipant · PT
UNIVERSITEIT VAN AMSTERDAMparticipant · NL
UNIVERSIDAD DE SEVILLAparticipant · ES

How to reach the team

Contact NTNU (Norges Teknisk-Naturvitenskapelige Universitet) in Norway — use Google AI Search to find the principal investigator's contact details.

Order a report on this consortium See NORGES TEKNISK-NATURVITENSKAPELIGE UNIVERSITET NTNU profile →

Next steps

Talk to the team behind this work.

Want to explore licensing the child speech recognition prototype or connect with the e-LADDA research team? SciTransfer can arrange a direct introduction and help evaluate fit for your product roadmap.

Order a report on this topic More in Digital & ICT