Language disorders are the most common developmental issue among children of preschool and elementary school age. Difficulties in language acquisition have far-reaching consequences, as 88% of affected children do not achieve the academic progress expected for their age group. In the long term, this can impair educational opportunities and social participation. It is therefore important to identify developmental risks early on to provide affected children with targeted support. At the same time, daycare centers, schools and therapeutic facilities face growing demands: Educational specialists are already working under significant time pressure and, in many places, are understaffed.

Digital technologies reduce professionals’ workload—leaving more time for what matters most

“Our technologies automatically analyze children’s spontaneous speech and reveal potential developmental difficulties, even at an early stage of language development,” explains Laura Tuschen, Head of the Assistive Speech and Language Analysis Group at Fraunhofer IDMT.

To this end, Laura Tuschen and her interdisciplinary team at Fraunhofer IDMT in Oldenburg are developing modern methods for automated analysis of children’s speech that analyze acoustic speech signals, transcribe them into text, and evaluate the results. Children’s speech poses unique challenges for automated analysis systems: Pronunciation, grammar, and speech patterns change with age and vary greatly from child to child. A key focus of development is analyzing spontaneous speech in everyday contexts to assess skills in natural situations. The software design and integrability of the developed methods enable a privacy-compliant and nuanced evaluation of children’s speech in natural communication situations.

It is important to note that digital speech and language analyses do not replace professional diagnosis. Instead, they are intended to support educational and therapeutic specialists by partially automating time-consuming routine tasks and systematizing observations. Professional assessment and the resulting decisions on support measures remain in the hands of experts.

Applications for recurring routine tasks and standardization

The researchers see potential applications for digital speech and language analysis in monitoring language development during everyday daycare activities, screening basic language skills before children enter school, and providing digital support and therapy services. Assessing and supporting children’s language development is complex: It requires differentiated professional expertise, time-intensive observation and documentation, and the availability of specialized professionals.

The current political debate also emphasizes the importance of early childhood education. The draft “Startchancen- und Qualitätsentwicklungsgesetz” (Act on Equal Opportunities and Quality Development) aims to strengthen the quality of child day care. Language support is central to this—with resulting additional requirements for documentation, support services, and monitoring of mandatory quality standards.

This creates new opportunities for providers of digital education, diagnostics, therapy and documentation solutions, as well as for service providers and public-sector stakeholders that are in need of scalable language screening. Integrating scientifically validated digital speech and language analysis into practical screening solutions can help identify difficulties at an early stage and facilitate access to support and therapy measures. It can also enable language development to be documented and tracked systematically.

For this purpose, Fraunhofer IDMT has developed and tested specific solutions for the preschool sector in various research and development projects and successfully evaluated them under real-world conditions.

These methods can be adapted as technological building blocks to different areas of application and integrated into digital diagnostic, documentation, support, or therapy solutions. To this end, Fraunhofer IDMT is seeking partners to jointly develop, test, and transfer specific applications into practice.

Practical development in collaboration with professionals

Researchers from Fraunhofer IDMT will discuss how digital speech technologies can be meaningfully integrated into practice at the 36^th National Congress of the “Deutsche Gesellschaft für Sprachheilpädagogik (dgs) e.V” (German Association for Speech and Language Therapy and Pedagogy), to be held September 24–26, 2026, in Oldenburg. On September 24, from 3:00 to 5:00 p.m., they will present current developments from the IDEAS and AuThenTo research projects in a workshop and, together with professionals, develop requirements for practical use in daycare centers, schools, and special education facilities.

The focus will be on questions regarding specific application needs, necessary framework conditions, data protection requirements, and acceptance among children, parents, guardians, and educational professionals.