BesserLesen: Reading promotion through app and AI-supported speech recognition Mobile application to promote reading skills of children
Technische Universität Braunschweig, in association with Ludwig-Maximilians-Universität München (LMU) and Technical University of Munich (TUM), has obtained funding of 1.317 million euros from the Federal Ministry of Education and Research for the BesserLesen (better reading) validation project. The aim of the joint project is to provide a demonstrative app that is embedded in an overall didactic concept for cooperative reading promotion for children of primary school age. Such an application is not yet available for the German language.
An AI-based automatic pronunciation assessment is used, which enables learning at the individual learning pace and is intended to contribute to improving children’s reading skills through targeted feedback. Joint project manager Professor Tim Fingscheidt from the Institute for Communications Technology at TU Braunschweig emphasises the importance of the many years of preliminary work: “The automatic speech recognition already developed at TU Braunschweig is the basis. This will be combined with algorithms for evaluating pronunciation quality during reading as part of the project.” TU Braunschweig’s share of the BMBF grant is 673,000 euros.
Pronunciation quality in reading primarily includes sound formation, e.g. whether the double sound “eu” in the word “Leute” (people) is pronounced correctly. Intonation, i.e. the melody of speech, is also examined. At the end of a question sentence, for example, the voice should be raised. Finally, the reading flow or speaking rate is automatically measured.
This core technology unfolds via the app within the framework of a cooperative reading promotion concept that is to be implemented and evaluated at primary schools. The head of the sub-project at LMU Munich, Prof. Uta Hauck-Thum, focuses on a changed understanding of teaching and learning in the digital world: “We are not only concerned with developing technologies further, but also teaching as such.” The app is also designed for independent learning at home.
Findings from the project can be used to reduce educational inequalities based on origin through the use of digital media. PD Dr. Jana Heinz heads the sub-project at the Technical University of Munich. She emphasises that the app is intended to improve equal opportunities and educational participation of pupils. The results will be made available to educators and teachers in training and further education.