If children with reading problems are not given targeted support, or if their needs are not recognized in primary school, this has a lasting impact on their written language skills. To help school students with their learning and at the same time relieve teaching staff, a digital solution for assessing German reading proficiency is to be developed within the LeseKind consortium. Children read texts aloud, which are then analyzed by means of automatic speech assessment, using the technologies for automatic speech recognition and analysis of Fraunhofer IDMT's Oldenburg Branch Hearing, Speech and Audio Technology HSA. In this way, specific parameters enable conclusions to be drawn about reading fluency and accuracy. An algorithm then assesses the children’s reading skills and rates their performance individually.

Regular and constructive feedback on individual reading competence is the foundation for reading success. Based on the assessment of children’s reading performance, individual training material can be offered. First of all, a reliable speech recognition system for children’s voices and read speech is to be developed in the project, followed by an assessment of reading proficiency with the aid of artificial intelligence. Together with the project partners, numerous different tasks and use cases will be developed in order to be able to record the competence levels along the reading development in childhood. Apart from integrating ethical and legal aspects, the aim is to conduct a comprehensive pilot study in the LeseKind project. The goal of the project is later integration into an eKidz.eu app. The digital tool is intended to be usable both for home use and in an educational context.

Further Information


News / 3.1.2024

Speech technologies that help us!

Laura Tuschen, head of Assistive Speech and Language Analysis group, gives an insight into the work of the research group.



The THERESIAH project aims to develop a new therapy system to train hearing and articulation of highly hearing-impaired persons.


Automatic speech analysis

Based on machine learning, speech and voice pathologies should be automatically evaluated and analysed in our SVAT project.


Connected Health

The work area Connected Health dedicates its research and development activities to sensor-based health systems that can be deployed flexibly and without clinical infrastructure.

The project LeseKind is funded by the German ministry of education and research (BMBF) and KMU-innovativ (01IS22009B).

Assistive Speech and Language Analysis

The development and utilisation of digital speech processing technologies can make an important contribution to individual capacity for verbal communication.