The Future of Education

Edition 14

Accepted Abstracts

Combining NLP, speech recognition, and indexing: an AI-based learning assistant for higher education

Tobias Schmohl, OWL University of Applied Sciences - Hamburg Centre for University Teaching & Learning (HUL) (Germany)

Kathrin Schelling, Management University of East Westphalia-Lippe (Germany)

Stefanie Go, Management University of East Westphalia-Lippe (Germany)

Carolin Freier, Technische Hochschule Nürnberg (Germany)

Marianne Hunger, Bavarian Center for Innovative Teaching (BayZiel) (Germany)

Franziska Hoffmann, Hochschule Ansbach (Germany)

Anne-Kathrin Helten, Evangelische Hochschule Nürnberg (Germany)

Florian Richter, Technische Hochschule Ingolstadt (Germany)


The project described in this article aims to develop HAnS (short for "Hochschul-Assistenz-System," or "assistance system for higher education"), an intelligent tutoring system (ITS) for higher education that will assist students from various disciplines in their pursuit of self-directed digital learning. From 2021 to 2025, this system will be developed and implemented by twelve cooperating German universities and research institutes, demonstrating the benefits of artificial intelligence (AI) and Big Data in higher education and, ideally, driving innovation in the field of technology-based learning. HAnS is based on pre-existing learning materials that will be processed via speech recognition. The platform will use AI to index the materials, allowing users to search for specific topics and to compile them automatically based on criteria such as lesson subject, language, medium, or required prior knowledge.


The proposed article explains how the HAnS project combines three educational and/or technical potentials to create the framework of an ITS which both students and teachers will be able to use to improve the effectiveness of self-study in higher education: (1) automatic transcription and indexing of audio-visual educational resources (e.g., lecture recordings, instructional videos, screencasts, podcasts), (2) personalized search for learning materials (including a recommender system), and (3) dynamic generation and gamification of exercises. To train the HAnS AI, we use authentic audio-visual teaching/learning materials (TLM) provided by teachers from various German universities. The intelligent system adaptively assembles these materials to generate dossiers on specific topics for self-study based on user information and educational guidelines embedded in the system. Additionally, the ITS will provide a more active form of support by automatically generating exercises based on predefined patterns and teaching materials, allowing learners to independently monitor their progress. 


Along with the (current and planned) features of the platform and the project’s goals, the article also outlines the development process, which is accompanied by an interdisciplinary team that will continuously re-evaluate and adapt the concept of HAnS over the course of several DBR cycles in order to closely match the ITS's features to the needs and learning habits of students in higher education. By collecting and analyzing educational research data (mixed methods design; primary and secondary research methods), we hope to derive implications for the system’s technical development and—in the long run—contribute to the integration of HAnS in everyday teaching and learning processes at the institutions participating in the project.


Back to the list


Reserved area

Media Partners:

Click BrownWalker Press logo for the International Academic and Industry Conference Event Calendar announcing scientific, academic and industry gatherings, online events, call for papers and journal articles
Pixel - Via Luigi Lanzi 12 - 50134 Firenze (FI) - VAT IT 05118710481
    Copyright © 2024 - All rights reserved

Privacy Policy